ProFusion2 — towards a Modular, Robust and Reliable Fusion Architecture for Automotive Environment Perception

Multisensor Data Fusion Strategies for Advanced Driver Assistance Systems

“…In this project, 16 objectdetecting sensors among four types of sensors are considered to be assembled on the test vehicle (Figure 3 The sensor network assembly consists of one Long Range Radar-LRR (2nd generation long range radar by Bosch) mounted for front monitoring (R L ), five Short Range Radar sensors (from M/ A-COM / Tyco Electronics) four of them in both sides and one in the front (R S ). Four laser scanners (from IBEO GmbH) with broad range and wide viewing angle, two of them in front and the others for rear side (L), three CMOS cameras INKANSC640PG by Aglaia GmbH (Amditis et al, 2006) two of them in side mirrors for rear and blind spot coverage and one in the middle of the front windscreen to face forward (C L ); A short range monocular camera for backside (C S ), and finally two ultrasonic on both sides (U). The placement of these object-detecting sensors and the main characteristics, such as detecting distance and the field of view, for these object-detection sensors can be seen in figure 3.…”

“…Passive safety systems such as seatbelts and airbags provide protection in the case of collision; more recently however, active safety systems have been introduced to help the driver avoid collisions in the first place. Nowadays, systems such as lane departure warning and rear-end collision avoidance have been introduced (Amditis et al, 2006); (Mammar et al, 2006). These active safety systems are required to interact much more with the driver than passive safety systems, creating a closed loop between driver, vehicle, and the environment.…”

Modelling Driver Behaviour in Automotive Environments

“…DVE real-time modelling includes the development of a perception system that extends existing or under development frameworks for safety of comfort applications, e.g. Tatschke et al (2006) in order to include driver's information to the perception architecture. The DVE module estimates the state of the DVE individual components and calculates level of risks as higher levels of abstraction for direct exploitation from the HMI component.…”

Multi level processing methodology for automotive applications

“…As our data structure builds the link between the different components of a safety system like the aforementioned one it provides a decoupling by means of generic interfaces between the system's modules. Thus our approach supports the independent development of components by relying only on the data structure's (input/output) interfaces and hence facilitates the implementation of interoperable (sensor data fusion) systems ( [10]). As our data management solution additionally offers data import and export routines an easy exchange of environmental data (from the perception data of sensor devices up to the description of tracked objects) is supported.…”